1996 年致股东信

BERKSHIRE HATHAWAY INC.
Chairman's Letter

伯克希尔·哈撒韦公司
董事长致股东信

To the Shareholders of Berkshire Hathaway Inc.:

致伯克希尔·哈撒韦公司股东：

Our gain in net worth during 1996 was $6.2 billion, or 36.1%. Per-share book value, however, grew by less, 31.8%, because the number of Berkshire shares increased: We issued stock in acquiring FlightSafety International and also sold new Class B shares.* Over the last 32 years (that is, since present management took over) per-share book value has grown from $19 to $19,011, or at a rate of 23.8% compounded annually.

1996 年我们的净资产增加了 62 亿美元，增幅为 36.1%。然而，每股账面价值增长较少，为 31.8%，原因是伯克希尔的股份数量增加了：我们在收购 FlightSafety International 时发行了股票，还出售了新的 B 类股票。* 在过去 32 年里（即自现任管理层接管以来），每股账面价值从 19 美元增长到 19,011 美元，年复合增长率为 23.8%。

• Each Class B share has an economic interest equal to 1/30th of that possessed by a Class A share, which is the new designation for the only stock that Berkshire had outstanding before May 1996. Throughout this report, we state all per-share figures in terms of "Class A equivalents," which are the sum of the Class A shares outstanding and 1/30th of the Class B shares outstanding.

• 每股 B 类股份的经济权益相当于 A 类股份的 1/30。A 类是伯克希尔在 1996 年 5 月之前唯一流通股的新名称。在本报告中，所有每股数据均以“A 类等值股份”表示，即已发行 A 类股份数量加上已发行 B 类股份数量的 1/30。

For technical reasons, we have restated our 1995 financial statements, a matter that requires me to present one of my less-than-thrilling explanations of accounting arcana. I'll make it brief.

由于技术原因，我们重述了 1995 年的财务报表，这需要我对会计晦涩之处做一个不那么激动人心的解释。我会尽量简短。

The restatement was required because GEICO became a wholly-owned subsidiary of Berkshire on January 2, 1996, whereas it was previously classified as an investment. From an economic viewpoint - taking into account major tax efficiencies and other benefits we gained - the value of the 51% of GEICO we owned at year-end 1995 increased significantly when we acquired the remaining 49% of the company two days later. Accounting rules applicable to this type of "step acquisition," however, required us to write down the value of our 51% at the time we moved to 100%. That writedown - which also, of course, reduced book value - amounted to $478.4 million. As a result, we now carry our original 51% of GEICO at a value that is both lower than its market value at the time we purchased the remaining 49% of the company and lower than the value at which we carry that 49% itself.

重述是因为 GEICO 在 1996 年 1 月 2 日成为伯克希尔的全资子公司，而此前它被归类为一项投资。从经济角度来看——考虑到我们获得的主要税收效率和其他好处——我们在 1995 年底拥有的 51% GEICO 股份的价值，在我们两天后收购剩余 49% 股份时显著增加。然而，适用于此类“分步收购”的会计准则要求我们在持股达到 100% 时减记那 51% 的价值。这次减记（当然也减少了账面价值）金额为 4.784 亿美元。结果，我们现在对最初 51% GEICO 的入账价值，既低于我们在购买剩余 49% 时它的市场价值，也低于我们对那 49% 本身的入账价值。

There is an offset, however, to the reduction in book value I have just described: Twice during 1996 we issued Berkshire shares at a premium to book value, first in May when we sold the B shares for cash and again in December when we used both A and B shares as part-payment for FlightSafety. In total, the three non-operational items affecting book value contributed less than one percentage point to our 31.8% per-share gain last year.

然而，对于我刚才描述的账面价值减少，有一个抵消项：1996 年期间我们有两次以高于账面价值的价格发行伯克希尔股票，第一次是在 5 月发行 B 股换取现金，第二次是在 12 月同时使用 A 股和 B 股作为收购 FlightSafety 的部分对价。总的来说，影响账面价值的三个非经营性项目对我们去年 31.8% 的每股收益增长贡献了不到一个百分点。

I dwell on this rise in per-share book value because it roughly indicates our economic progress during the year. But, as Charlie Munger, Berkshire's Vice Chairman, and I have repeatedly told you, what counts at Berkshire is intrinsic value, not book value. The last time you got that message from us was in the Owner's Manual, sent to you in June after we issued the Class B shares. In that manual, we not only defined certain key terms - such as intrinsic value - but also set forth our economic principles.

我之所以详细讨论每股账面价值的增长，是因为它大致反映了我们当年的经济进展。但是，正如伯克希尔副董事长查理·芒格和我一再告诉你们的，在伯克希尔真正重要的是内在价值，而不是账面价值。你们上次从我们这里得到这个信息是在我们发行 B 股后于 6 月寄给你们的《所有者手册》中。在那本手册中，我们不仅定义了某些关键术语——如内在价值——还阐述了我们经济原则。

For many years, we have listed these principles in the front of our annual report, but in this report, on pages 58 to 67, we reproduce the entire Owner's Manual. In this letter, we will occasionally refer to the manual so that we can avoid repeating certain definitions and explanations. For example, if you wish to brush up on "intrinsic value," see pages 64 and 65.

多年来，我们一直将这些原则列在年报的前面，但在本报告中，我们在第 58 至 67 页重印了整本《所有者手册》。在这封信中，我们会偶尔提及该手册，以避免重复某些定义和解释。例如，如果你想重温“内在价值”，请参见第 64 和 65 页。

Last year, for the first time, we supplied you with a table that Charlie and I believe will help anyone trying to estimate Berkshire's intrinsic value. In the updated version of that table, which follows, we trace two key indices of value. The first column lists our per-share ownership of investments (including cash and equivalents) and the second column shows our per-share earnings from Berkshire's operating businesses before taxes and purchase-accounting adjustments but after all interest and corporate overhead expenses. The operating-earnings column excludes all dividends, interest and capital gains that we realized from the investments presented in the first column. In effect, the two columns show what Berkshire would have reported had it been broken into two parts.

去年，我们首次向你们提供了一张表格，查理和我相信它能帮助任何人估算伯克希尔的内在价值。在下面更新的表格中，我们追踪了两个关键的价值指标。第一列列出了我们每股持有的投资（包括现金及等价物），第二列显示了伯克希尔运营业务在扣除所有利息和公司管理费用后、但在税和购买会计调整之前的每股收益。营业收益列排除了我们从第一列所示投资中实现的所有股息、利息和资本收益。实际上，这两列展示了如果伯克希尔被拆分为两部分将会报告的情况。

As the table tells you, our investments per share increased in 1996 by 29.0% and our non-investment earnings grew by 63.2%. Our goal is to keep the numbers in both columns moving ahead at a reasonable (or, better yet, unreasonable) pace.

正如表格所示，1996 年我们的每股投资增长了 29.0%，非投资收益增长了 63.2%。我们的目标是让两列的数字以合理（或者更好，不合理）的速度向前发展。

Our expectations, however, are tempered by two realities. First, our past rates of growth cannot be matched nor even approached: Berkshire's equity capital is now large - in fact, fewer than ten businesses in America have capital larger - and an abundance of funds tends to dampen returns. Second, whatever our rate of progress, it will not be smooth: Year-to-year moves in the first column of the table above will be influenced in a major way by fluctuations in securities markets; the figures in the second column will be affected by wide swings in the profitability of our catastrophe-reinsurance business.

然而，我们的预期受到两个现实情况的制约。第一，我们过去的增长率无法被匹配甚至接近：伯克希尔的股本现在已经很大——事实上，美国只有不到十家企业的资本比我们更大——而资金过剩往往会抑制回报。第二，无论我们的进展速度如何，它都不会是平稳的：上表第一列的逐年变动将受到证券市场波动的重大影响；第二列的数字将受到我们巨灾再保险业务盈利能力大幅波动的影响。

In the table, the donations made pursuant to our shareholder-designated contributions program are charged against the second column, though we view them as a shareholder benefit rather than as an expense. All other corporate expenses are also charged against the second column. These costs may be lower than those of any other large American corporation: Our after-tax headquarters expense amounts to less than two basis points (1/50th of 1%) measured against net worth. Even so, Charlie used to think this expense percentage outrageously high, blaming it on my use of Berkshire's corporate jet, The Indefensible. But Charlie has recently experienced a "counter-revelation": With our purchase of FlightSafety, whose major activity is the training of corporate pilots, he now rhapsodizes at the mere mention of jets.

在表格中，根据股东指定捐赠计划进行的捐赠被计入第二列，尽管我们将其视为股东福利而非费用。所有其他公司费用也被计入第二列。这些成本可能低于任何其他美国大公司：按净资产计算，我们的税后总部费用不到两个基点（万分之一）。即便如此，查理曾经认为这个费用比例高得离谱，归咎于我使用伯克希尔的公务机“无可辩驳号”。但查理最近经历了一次“反向启示”：随着我们收购了主要业务是培训公司飞行员的 FlightSafety，他现在一提到喷气机就眉飞色舞。

Seriously, costs matter. For example, equity mutual funds incur corporate expenses - largely payments to the funds' managers - that average about 100 basis points, a levy likely to cut the returns their investors earn by 10% or more over time. Charlie and I make no promises about Berkshire's results. We do promise you, however, that virtually all of the gains Berkshire makes will end up with shareholders. We are here to make money with you, not off you.

严肃地说，成本很重要。例如，股票共同基金产生的公司费用——主要是支付给基金经理的——平均约为 100 个基点，这项费用很可能会随着时间的推移将投资者的回报削减 10% 或更多。查理和我对伯克希尔的业绩不做任何承诺。但我们可以向你保证，伯克希尔产生的几乎所有收益最终都会归股东所有。我们在这里是与你一起赚钱，而不是从你身上赚钱。

In last year's letter, with Berkshire shares selling at $36,000, I told you: (1) Berkshire's gain in market value in recent years had outstripped its gain in intrinsic value, even though the latter gain had been highly satisfactory; (2) that kind of overperformance could not continue indefinitely; (3) Charlie and I did not at that moment consider Berkshire to be undervalued.

在去年的信中，当伯克希尔股价为 36,000 美元时，我告诉你：(1) 尽管内在价值增长非常令人满意，但伯克希尔近年来的市场价值增长超过了内在价值增长；(2) 这种跑赢不可能无限期持续下去；(3) 查理和我当时不认为伯克希尔被低估。

Since I set down those cautions, Berkshire's intrinsic value has increased very significantly - aided in a major way by a stunning performance at GEICO that I will tell you more about later - while the market price of our shares has changed little. This, of course, means that in 1996 Berkshire's stock underperformed the business. Consequently, today's price/value relationship is both much different from what it was a year ago and, as Charlie and I see it, more appropriate.

自从我写下这些提醒以来，伯克希尔的内在价值已经显著增加——很大程度上得益于 GEICO 惊人的表现（我稍后会详细告诉你）——而我们股票的市场价格几乎没有变化。这当然意味着 1996 年伯克希尔的股票表现落后于业务。因此，如今的价格/价值关系与一年前大不相同，而且在查理和我看来，更为合适。

Over time, the aggregate gains made by Berkshire shareholders must of necessity match the business gains of the company. When the stock temporarily overperforms or underperforms the business, a limited number of shareholders - either sellers or buyers - receive outsized benefits at the expense of those they trade with. Generally, the sophisticated have an edge over the innocents in this game.

随着时间的推移，伯克希尔股东的总收益必然与公司的业务收益相匹配。当股票暂时跑赢或跑输业务时，少数股东——无论是卖出方还是买入方——会以与他们交易的人为代价获得超额的收益。通常，在这场游戏中，老练者比天真者更有优势。

Though our primary goal is to maximize the amount that our shareholders, in total, reap from their ownership of Berkshire, we wish also to minimize the benefits going to some shareholders at the expense of others. These are goals we would have were we managing a family partnership, and we believe they make equal sense for the manager of a public company. In a partnership, fairness requires that partnership interests be valued equitably when partners enter or exit; in a public company, fairness prevails when market price and intrinsic value are in sync. Obviously, they won't always meet that ideal, but a manager - by his policies and communications - can do much to foster equity.

虽然我们的首要目标是最大化所有股东从持有伯克希尔中获得的总收益，但我们也希望尽量减少以牺牲其他股东为代价而流向部分股东的利益。如果我们管理的是家族合伙企业，这也是我们的目标，而且我们认为这对上市公司的管理者同样有意义。在合伙企业中，公平要求合伙人在进入或退出时，合伙权益得到公平估值；在上市公司中，当市场价格与内在价值同步时，公平就实现了。显然，它们不会总是达到这种理想状态，但管理者——通过他的政策和沟通——可以大有作为来促进公平。

Of course, the longer a shareholder holds his shares, the more bearing Berkshire's business results will have on his financial experience - and the less it will matter what premium or discount to intrinsic value prevails when he buys and sells his stock. That's one reason we hope to attract owners with long-term horizons. Overall, I think we have succeeded in that pursuit. Berkshire probably ranks number one among large American corporations in the percentage of its shares held by owners with a long-term view.

当然，股东持股时间越长，伯克希尔的经营业绩对其财务体验的影响就越大——而他在买卖股票时相对于内在价值的溢价或折价的影响就越小。这就是我们希望吸引具有长期视野的所有者的原因之一。总体而言，我认为我们在这方面取得了成功。在大市值美国公司中，伯克希尔长期持股者的持股比例可能排名第一。

We made two acquisitions in 1996, both possessing exactly the qualities we seek - excellent business economics and an outstanding manager.

1996 年我们进行了两次收购，两者都完全具备我们寻求的品质——卓越的商业经济特性和杰出的管理者。

The first acquisition was Kansas Bankers Surety (KBS), an insurance company whose name describes its specialty. The company, which does business in 22 states, has an extraordinary underwriting record, achieved through the efforts of Don Towle, an extraordinary manager. Don has developed first-hand relationships with hundreds of bankers and knows every detail of his operation. He thinks of himself as running a company that is "his," an attitude we treasure at Berkshire. Because of its relatively small size, we placed KBS with Wesco, our 80%-owned subsidiary, which has wanted to expand its insurance operations.

第一笔收购是 Kansas Bankers Surety (KBS)，一家保险公司，其名称说明了其业务专长。该公司在 22 个州开展业务，拥有非凡的承保记录，这得益于非凡的管理者 Don Towle 的努力。Don 与数百名银行家建立了第一手关系，并且了解其运营的每一个细节。他认为自己经营着一家“他的”公司，这种态度是我们在伯克希尔珍视的。由于规模相对较小，我们将 KBS 归入 Wesco，这是我们持股 80% 的子公司，该公司一直希望扩大其保险业务。

You might be interested in the carefully-crafted and sophisticated acquisition strategy that allowed Berkshire to nab this deal. Early in 1996 I was invited to the 40th birthday party of my nephew's wife, Jane Rogers. My taste for social events being low, I immediately, and in my standard, gracious way, began to invent reasons for skipping the event. The party planners then countered brilliantly by offering me a seat next to a man I always enjoy, Jane's dad, Roy Dinsdale - so I went.

你可能会对伯克希尔成功达成这笔交易的精巧而复杂的收购策略感兴趣。1996 年初，我被邀请参加我侄子的妻子 Jane Rogers 的 40 岁生日派对。我对社交活动的兴趣不高，于是我立即以我一贯优雅的方式开始编造跳过活动的理由。然后派对策划者做出了巧妙的回应，安排我坐在我一直很欣赏的人——Jane 的父亲 Roy Dinsdale 旁边——于是我去了。

The party took place on January 26. Though the music was loud - Why must bands play as if they will be paid by the decibel? - I just managed to hear Roy say he'd come from a directors meeting at Kansas Bankers Surety, a company I'd always admired. I shouted back that he should let me know if it ever became available for purchase.

派对在 1 月 26 日举行。虽然音乐很吵——为什么乐队非要像按分贝收费一样演奏？——我勉强听到 Roy 说他刚从 Kansas Bankers Surety 的董事会上下来，这家公司我一直很欣赏。我大声回答说，如果这家公司有出售意向，请告诉我。

On February 12, I got the following letter from Roy: "Dear Warren: Enclosed is the annual financial information on Kansas Bankers Surety. This is the company that we talked about at Janie's party. If I can be of any further help, please let me know." On February 13, I told Roy we would pay $75 million for the company - and before long we had a deal. I'm now scheming to get invited to Jane's next party.

2 月 12 日，我收到了 Roy 的来信：“亲爱的沃伦：随信附上 Kansas Bankers Surety 的年度财务信息。这就是我们在 Janie 派对上谈到的那家公司。如果我能提供任何进一步的帮助，请告诉我。”2 月 13 日，我告诉 Roy，我们将支付 7500 万美元收购这家公司——不久我们就达成了协议。我现在正在策划如何得到 Jane 下个派对的邀请。

Our other acquisition in 1996 - FlightSafety International, the world's leader in the training of pilots - was far larger, at about $1.5 billion, but had an equally serendipitous origin. The heroes of this story are first, Richard Sercer, a Tucson aviation consultant, and second, his wife, Alma Murphy, an ophthalmology graduate of Harvard Medical School, who in 1990 wore down her husband's reluctance and got him to buy Berkshire stock. Since then, the two have attended all our Annual Meetings, but I didn't get to know them personally.

我们 1996 年的另一项收购——FlightSafety International，全球飞行员培训领域的领导者——规模大得多，约 15 亿美元，但其起源同样具有偶然性。这个故事的主角首先是 Richard Sercer，一位图森市的航空顾问；其次是他的妻子 Alma Murphy，哈佛医学院眼科毕业生，她在 1990 年磨掉了丈夫的犹豫，让他买了伯克希尔的股票。从那以后，他们两人参加了我们的每一次年会，但我并没有亲自认识他们。

Fortunately, Richard had also been a long-time shareholder of FlightSafety, and it occurred to him last year that the two companies would make a good fit. He knew our acquisition criteria, and he thought that Al Ueltschi, FlightSafety's 79-year-old CEO, might want to make a deal that would both give him a home for his company and a security in payment that he would feel comfortable owning throughout his lifetime. So in July, Richard wrote Bob Denham, CEO of Salomon Inc, suggesting that he explore the possibility of a merger.

幸运的是，Richard 也一直是 FlightSafety 的长期股东，去年他突然想到这两家公司会很合适。他知道我们的收购标准，他认为 FlightSafety 79 岁的 CEO Al Ueltschi 可能想要达成一笔交易，既能为他的公司找个归宿，又能得到一种他终身持有都感到安心的支付证券。于是在 7 月，Richard 写信给所罗门公司的 CEO Bob Denham，建议他探讨合并的可能性。

Bob took it from there, and on September 18, Al and I met in New York. I had long been familiar with FlightSafety's business, and in about 60 seconds I knew that Al was exactly our kind of manager. A month later, we had a contract. Because Charlie and I wished to minimize the issuance of Berkshire shares, the transaction we structured gave FlightSafety shareholders a choice of cash or stock but carried terms that encouraged those who were tax-indifferent to take cash. This nudge led to about 51% of FlightSafety's shares being exchanged for cash, 41% for Berkshire A and 8% for Berkshire B.

Bob 接手了这件事，9 月 18 日，Al 和我在纽约会面。我早就熟悉 FlightSafety 的业务，大约 60 秒内我就知道 Al 正是我们想要的那种经理人。一个月后，我们签了合同。由于查理和我希望尽量减少伯克希尔股票的发行量，我们构建的交易给了 FlightSafety 股东现金或股票的选择权，但条款鼓励那些对税收无所谓的人选择现金。这种引导使得约 51% 的 FlightSafety 股份换成了现金，41% 换成了伯克希尔 A 股，8% 换成了伯克希尔 B 股。

Al has had a lifelong love affair with aviation and actually piloted Charles Lindbergh. After a barnstorming career in the 1930s, he began working for Juan Trippe, Pan Am's legendary chief. In 1951, while still at Pan Am, Al founded FlightSafety, subsequently building it into a simulator manufacturer and a worldwide trainer of pilots (single-engine, helicopter, jet and marine). The company operates in 41 locations, outfitted with 175 simulators of planes ranging from the very small, such as Cessna 210s, to Boeing 747s. Simulators are not cheap - they can cost as much as $19 million - so this business, unlike many of our operations, is capital intensive. About half of the company's revenues are derived from the training of corporate pilots, with most of the balance coming from airlines and the military.

Al 一生热爱航空，实际上他曾为查尔斯·林德伯格驾驶过飞机。在 1930 年代从事巡回飞行表演后，他开始为泛美航空公司传奇领袖胡安·特里普工作。1951 年，仍在泛美期间，Al 创立了 FlightSafety，随后将其打造成模拟器制造商和全球飞行员培训商（单引擎、直升机、喷气机和海上飞机）。该公司在 41 个地点运营，配备了 175 台模拟器，涵盖从塞斯纳 210 等非常小的飞机到波音 747 等各种机型。模拟器并不便宜——成本可能高达 1900 万美元——所以这项业务与我们许多业务不同，是资本密集型的。公司约一半的收入来自公司飞行员的培训，其余大部分来自航空公司和军方。

Al may be 79, but he looks and acts about 55. He will run operations just as he has in the past: We never fool with success. I have told him that though we don't believe in splitting Berkshire stock, we will split his age 2-for-1 when he hits 100.

Al 可能已经 79 岁了，但他的外表和行为都像 55 岁。他会像过去一样经营业务：我们从不折腾成功。我告诉他，虽然我们不相信拆分公司股票，但等他到 100 岁时，我们会把他的年龄做 2 比 1 的分拆。

An observer might conclude from our hiring practices that Charlie and I were traumatized early in life by an EEOC bulletin on age discrimination. The real explanation, however, is self-interest: It's difficult to teach a new dog old tricks. The many Berkshire managers who are past 70 hit home runs today at the same pace that long ago gave them reputations as young slugging sensations. Therefore, to get a job with us, just employ the tactic of the 76-year-old who persuaded a dazzling beauty of 25 to marry him. "How did you ever get her to accept?" asked his envious contemporaries. The comeback: "I told her I was 86."

观察者可能会从我们的雇佣实践中得出结论，认为查理和我在早年受到过 EEOC 关于年龄歧视公告的创伤。然而，真正的解释是自利：教新狗老把戏很难。伯克希尔许多年过 70 的经理人今天仍以当年让他们赢得年轻重炮手声誉的速度打出本垒打。因此，要得到我们这里的工作，只需采用那位 76 岁老人的策略，他说服了一位 25 岁的惊艳美女嫁给他。“你是怎么让她答应的？”羡慕的同龄人问道。回答：“我告诉她我 86 岁了。”

And now we pause for our usual commercial: If you own a large business with good economic characteristics and wish to become associated with an exceptional collection of businesses having similar characteristics, Berkshire may well be the home you seek. Our requirements are set forth on page 21. If your company meets them - and if I fail to make the next birthday party you attend - give me a call.

现在我们暂停一下，插播我们惯常的广告：如果你拥有一家具有良好经济特性的大型企业，并希望与一群具有类似特性的卓越企业为伍，伯克希尔很可能就是你寻找的家。我们的要求在第 21 页列出。如果你的公司符合条件——而且如果我未能出席你参加的下一个生日派对——请给我打电话。

Our insurance business was terrific in 1996. In both primary insurance, where GEICO is our main unit, and in our "super-cat" reinsurance business, results were outstanding.

1996 年我们的保险业务非常出色。在 GEICO 作为主要部门的初级保险业务和我们的“超级巨灾”再保险业务中，业绩都非常突出。

As we've explained in past reports, what counts in our insurance business is, first, the amount of "float" we generate and, second, its cost to us. These are matters that are important for you to understand because float is a major component of Berkshire's intrinsic value that is not reflected in book value.

正如我们在过去报告中解释过的，我们保险业务的关键首先是所产生的“浮存金”数量，其次是其成本。这些事项对你理解很重要，因为浮存金是伯克希尔内在价值的一个主要组成部分，而它并没有反映在账面价值中。

To begin with, float is money we hold but don't own. In an insurance operation, float arises because premiums are received before losses are paid. Secondly, the premiums that an insurer takes in typically do not cover the losses and expenses it eventually must pay. That leaves it running an "underwriting loss," which is the cost of float. An insurance business has value if its cost of float over time is less than the cost the company would otherwise incur to obtain funds. But the business is an albatross if the cost of its float is higher than market rates for money.

首先，浮存金是我们持有但不拥有的钱。在保险业务中，浮存金产生的原因是保费在损失赔付之前收到。其次，保险公司收取的保费通常无法覆盖其最终必须支付的损失和费用。这就导致了“承保亏损”，即浮存金的成本。如果一家保险公司的浮存金成本随着时间的推移低于公司以其他方式获取资金的成本，那么该业务就有价值。但如果其浮存金成本高于市场利率，那么该业务就是一个累赘。

As the numbers in the following table show, Berkshire's insurance business has been a huge winner. For the table, we have calculated our float - which we generate in large amounts relative to our premium volume - by adding loss reserves, loss adjustment reserves, funds held under reinsurance assumed and unearned premium reserves, and then subtracting agents' balances, prepaid acquisition costs, prepaid taxes and deferred charges applicable to assumed reinsurance. Our cost of float is determined by our underwriting loss or profit. In those years when we have had an underwriting profit, such as the last four, our cost of float has been negative. In effect, we have been paid for holding money.

如下表数字所示，伯克希尔的保险业务一直是一个巨大的赢家。对于该表，我们通过加总损失准备金、损失调整准备金、分保业务持有的资金和未赚保费准备金，再减去代理人余额、预付 acquisition 成本、预付税款和适用于分保业务的递延费用，来计算我们的浮存金——相对于保费规模，我们产生了大量的浮存金。我们的浮存金成本由我们的承保损失或利润决定。在那些我们有承保利润的年份（例如过去四年），我们的浮存金成本为负。实际上，我们持有 money 还能获得报酬。

Since 1967, when we entered the insurance business, our float has grown at an annual compounded rate of 22.3%. In more years than not, our cost of funds has been less than nothing. This access to "free" money has boosted Berkshire's performance in a major way. Moreover, our acquisition of GEICO materially increases the probability that we can continue to obtain "free" funds in increasing amounts.

自 1967 年我们进入保险业务以来，我们的浮存金以年复合增长率 22.3% 增长。在大多数年份，我们的资金成本都低于零。这种获得“免费”资金的途径极大地提升了伯克希尔的业绩。此外，我们收购 GEICO 大大增加了我们继续以增加的数量获得“免费”资金的可能性。

As in the past three years, we once again stress that the good results we are reporting for Berkshire stem in part from our super-cat business having a lucky year. In this operation, we sell policies that insurance and reinsurance companies buy to protect themselves from the effects of mega-catastrophes. Since truly major catastrophes are rare occurrences, our super-cat business can be expected to show large profits in most years - and to record a huge loss occasionally. In other words, the attractiveness of our super-cat business will take a great many years to measure. What you must understand, however, is that a truly terrible year in the super-cat business is not a possibility - it's a certainty. The only question is when it will come.

与过去三年一样，我们再次强调，我们报告的伯克希尔良好业绩部分来源于我们的超级巨灾业务度过了一个幸运的年份。在这项业务中，我们出售保单给保险公司和再保险公司，以保护它们免受 mega-catastrophes 的影响。由于真正的大灾难很少发生，我们的超级巨灾业务在大多数年份预计会显示出 large profits——偶尔会 record 巨额损失。换句话说，我们超级巨灾业务的吸引力需要很多年才能衡量。然而，你必须理解的是，超级巨灾业务真正糟糕的年份不是一种可能性——而是一种 certainty。唯一的问题是它何时到来。

I emphasize this lugubrious point because I would not want you to panic and sell your Berkshire stock upon hearing that some large catastrophe had cost us a significant amount. If you would tend to react that way, you should not own Berkshire shares now, just as you should entirely avoid owning stocks if a crashing market would lead you to panic and sell. Selling fine businesses on "scary" news is usually a bad decision. (Robert Woodruff, the business genius who built Coca-Cola over many decades and who owned a huge position in the company, was once asked when it might be a good time to sell Coke stock. Woodruff had a simple answer: "I don't know. I've never sold any.")

我强调这个令人沮丧的观点，因为我不希望你在听到某场大灾难让我们损失惨重时就 panic 并卖出伯克希尔的股票。如果你倾向于那样反应，你现在就不应该持有伯克希尔的股票，就像如果市场 crash 会导致你恐慌卖出，你就应该完全避免持有股票一样。基于“可怕”的消息卖出优质企业通常是一个糟糕的决定。（Robert Woodruff 是花费几十年打造可口可乐的商业天才，并持有该公司巨额头寸。曾有人问他什么时候是卖出可口可乐股票的好时机。Woodruff 的回答很简单：“我不知道。我从来没卖过。”）

In our super-cat operation, our customers are insurers that are exposed to major earnings volatility and that wish to reduce it. The product we sell - for what we hope is an appropriate price - is our willingness to shift that volatility to our own books. Gyrations in Berkshire's earnings don't bother us in the least: Charlie and I would much rather earn a lumpy 15% over time than a smooth 12%. (After all, our earnings swing wildly on a daily and weekly basis - why should we demand that smoothness accompany each orbit that the earth makes of the sun?) We are most comfortable with that thinking, however, when we have shareholder/partners who can also accept volatility, and that's why we regularly repeat our cautions.

在我们的超级巨灾业务中，我们的客户是那些面临重大收益波动并希望减少波动的保险公司。我们出售的产品——希望以合适的价格——是我们愿意将这种波动转移到我们自己账面上的意愿。伯克希尔收益的波动丝毫不会困扰我们：查理和我宁愿随着时间的推移获得不稳定的 15% 回报，也不愿获得平稳的 12% 回报。（毕竟，我们的收益每天都在每周都在剧烈波动——为什么我们要要求伴随着地球绕太阳的每一圈都有平稳？）然而，只有当我们的股东/合伙人也能接受波动性时，我们对这种想法才最为舒适，这就是为什么我们定期重复我们的提醒。

We took on some major super-cat exposures during 1996. At mid-year we wrote a contract with Allstate that covers Florida hurricanes, and though there are no definitive records that would allow us to prove this point, we believe that to have then been the largest single catastrophe risk ever assumed by one company for its own account. Later in the year, however, we wrote a policy for the California Earthquake Authority that goes into effect on April 1, 1997, and that exposes us to a loss more than twice that possible under the Florida contract. Again we retained all the risk for our own account. Large as these coverages are, Berkshire's after-tax "worst-case" loss from a true mega-catastrophe is probably no more than $600 million, which is less than 3% of our book value and 1.5% of our market value. To gain some perspective on this exposure, look at the table on page 2 and note the much greater volatility that security markets have delivered us.

1996 年我们承担了一些重大的超级巨灾风险敞口。年中我们与 Allstate 签订了一份覆盖佛罗里达飓风的合同，虽然没有确切的记录能让我们证明这一点，但我们相信这是当时一家公司为自有账户承担的最大的单次巨灾风险。然而，那年晚些时候，我们为加州地震局承保了一份保单，该保单于 1997 年 4 月 1 日生效，其所暴露给我们的损失可能是佛罗里达合同下可能损失的两倍多。我们再次为自有账户保留了所有风险。尽管这些覆盖范围很大，但伯克希尔从一次真正的 mega-catastrophe 中获得的税后“最坏情况”损失可能不超过 6 亿美元，不到我们账面价值的 3% 和我们市值的 1.5%。为了更直观地理解这种风险敞口，请看第 2 页的表格，并注意证券市场给我们带来的更大的波动性。

In the super-cat business, we have three major competitive advantages. First, the parties buying reinsurance from us know that we both can and will pay under the most adverse of circumstances. Were a truly cataclysmic disaster to occur, it is not impossible that a financial panic would quickly follow. If that happened, there could well be respected reinsurers that would have difficulty paying at just the moment that their clients faced extraordinary needs. Indeed, one reason we never "lay off" part of the risks we insure is that we have reservations about our ability to collect from others when disaster strikes. When it's Berkshire promising, insureds know with certainty that they can collect promptly.

在超级巨灾业务中，我们有三大竞争优势。第一，向我们购买再保险的各方知道，在最不利的情况下，我们能够并且将会赔付。如果真的发生一场灾难性的大灾难，随之而来的很可能是金融恐慌。如果发生这种情况，很可能一些受人尊敬的再保险公司会恰恰在其客户面临特殊需求时难以赔付。事实上，我们从不“分保”我们所承保的部分风险的一个原因是，我们对灾难发生时从其他人那里收取赔付的能力持保留态度。当伯克希尔做出承诺时，被保险人确切地知道他们可以及时获得赔付。

Our second advantage - somewhat related - is subtle but important. After a mega-catastrophe, insurers might well find it difficult to obtain reinsurance even though their need for coverage would then be particularly great. At such a time, Berkshire would without question have very substantial capacity available - but it will naturally be our long-standing clients that have first call on it. That business reality has made major insurers and reinsurers throughout the world realize the desirability of doing business with us. Indeed, we are currently getting sizable "stand-by" fees from reinsurers that are simply nailing down their ability to get coverage from us should the market tighten.

我们的第二个优势——有点相关——微妙但很重要。在 mega-catastrophe 之后，保险公司可能会发现很难获得再保险，尽管那时他们对保障的需求特别大。在这样的时刻，伯克希尔毫无疑问会拥有非常可观的 capacity 可用——但自然是我们长期客户拥有优先使用权。这种商业现实使得世界各地的 major 保险公司和再保险公司意识到与我们做生意的可取性。事实上，我们目前正在从再保险公司那里获得可观的“ standby”费用，它们 simply 是为了确定在市场紧张时从我们这里获得保障的能力。

Our final competitive advantage is that we can provide dollar coverages of a size neither matched nor approached elsewhere in the industry. Insurers looking for huge covers know that a single call to Berkshire will produce a firm and immediate offering.

我们最后的竞争优势是，我们可以提供行业内其他公司无法比拟甚至无法接近的美元保险额。寻求巨额保障的保险公司知道，只要给伯克希尔打一个电话，就会得到一个 firm 和 immediate 的报价。

A few facts about our exposure to California earthquakes - our largest risk - seem in order. The Northridge quake of 1994 laid homeowners' losses on insurers that greatly exceeded what computer models had told them to expect. Yet the intensity of that quake was mild compared to the "worst-case" possibility for California. Understandably, insurers became - ahem - shaken and started contemplating a retreat from writing earthquake coverage into their homeowners' policies.

关于我们对加州地震风险敞口的一些事实——我们的最大风险——似乎是合适的。1994 年的北岭地震给保险公司带来的房主损失远远超过计算机模型告诉他们的预期。然而，与加州“最坏情况”的可能性相比，那次地震的强度还算温和。可以理解的是，保险公司变得——咳咳——动摇了，并开始考虑从给房主保单写入地震保障中退缩。

In a thoughtful response, Chuck Quackenbush, California's insurance commissioner, designed a new residential earthquake policy to be written by a state-sponsored insurer, The California Earthquake Authority. This entity, which went into operation on December 1, 1996, needed large layers of reinsurance - and that's where we came in. Berkshire's layer of approximately $1 billion will be called upon if the Authority's aggregate losses in the period ending March 31, 2001 exceed about $5 billion. (The press originally reported larger figures, but these would have applied only if all California insurers had entered into the arrangement; instead only 72% signed up.)

作为深思熟虑的回应，加州保险专员 Chuck Quackenbush 设计了一种新的住宅地震保单，由州政府支持的保险公司——加州地震局承保。该机构于 1996 年 12 月 1 日开始运作，需要大量的再保险保障——而这正是我们介入的地方。如果该机构在截至 2001 年 3 月 31 日的期间总损失超过约 50 亿美元，伯克希尔约 10 亿美元的保障层将被调用。（媒体最初报道的数字更大，但那仅适用于所有加州保险公司都加入该安排的情况；实际上只有 72% 签约了。）

So what are the true odds of our having to make a payout during the policy's term? We don't know - nor do we think computer models will help us, since we believe the precision they project is a chimera. In fact, such models can lull decision-makers into a false sense of security and thereby increase their chances of making a really huge mistake. We've already seen such debacles in both insurance and investments. Witness "portfolio insurance," whose destructive effects in the 1987 market crash led one wag to observe that it was the computers that should have been jumping out of windows.

那么，我们在保单有效期内需要赔付的真实 odds 是多少？我们不知道——我们也不认为计算机模型能帮到我们，因为我们相信它们所预测的精度是 chimera。事实上，此类模型会让决策者产生错误的安全感，从而增加他们犯下真正巨大错误的机会。我们在保险和投资领域都已经目睹过这样的失败。看看“投资组合保险”，它在 1987 年市场 crash 中的破坏性效果让一个 wag 评论说，应该是电脑跳楼才对。

Even if perfection in assessing risks is unattainable, insurers can underwrite sensibly. After all, you need not know a man's precise age to know that he is old enough to vote nor know his exact weight to recognize his need to diet. In insurance, it is essential to remember that virtually all surprises are unpleasant, and with that in mind we try to price our super-cat exposures so that about 90% of total premiums end up being eventually paid out in losses and expenses. Over time, we will find out how smart our pricing has been, but that will not be quickly. The super-cat business is just like the investment business in that it often takes a long time to find out whether you knew what you were doing.

即使无法达到评估风险的完美，保险公司也可以进行明智的承保。毕竟，你不需要知道一个人的确切年龄就知道他已经到了投票的年龄，也不需要知道他的确切体重就知道他需要节食。在保险业，必须记住， virtually 所有的 surprise 都是令人不快的，考虑到这一点，我们试图为我们超级巨灾风险敞口定价，使得总保费中约 90% 最终以损失和费用的形式支付出去。随着时间的推移，我们会发现我们的定价有多 smart，但这不会很快。超级巨灾业务就像投资业务一样，通常需要很长时间才能发现你是否知道自己在做什么。

What I can state with certainty, however, is that we have the best person in the world to run our super-cat business: Ajit Jain, whose value to Berkshire is simply enormous. In the reinsurance field, disastrous propositions abound. I know that because I personally embraced all too many of these in the 1970s and also because GEICO has a large runoff portfolio made up of foolish contracts written in the early-1980s, able though its then-management was. Ajit, I can assure you, won't make mistakes of this type.

然而，我可以确定地 state 的是，我们有世界上最好的人来经营我们的超级巨灾业务：Ajit Jain，他对伯克希尔的价值 simply enormous。在再保险领域，灾难性的 proposition 比比皆是。我知道这一点，因为我在 1970 年代 personally embraced 了太多这样的 proposition，也因为 GEICO 有一个庞大的业务组合，其中包含 1980 年代初期签订的愚蠢合同，尽管当时的管理层很能干。Ajit，我可以向你保证，不会犯这类错误。

I have mentioned that a mega-catastrophe might cause a catastrophe in the financial markets, a possibility that is unlikely but not far-fetched. Were the catastrophe a quake in California of sufficient magnitude to tap our coverage, we would almost certainly be damaged in other ways as well. For example, See's, Wells Fargo and Freddie Mac could be hit hard. All in all, though, we can handle this aggregation of exposures.

我已经提到过，一场 mega-catastrophe 可能会在金融市场引发一场 catastrophe，这种可能性不大但并非牵强。如果这场 catastrophe 是加州的强震，其规模足以动用我们的保障，那么几乎可以肯定我们也会在其他方面受到损害。例如，See's、富国银行和房地美可能会受到严重打击。不过，总的来说，我们可以处理这种风险敞口的聚集。

In this respect, as in others, we try to "reverse engineer" our future at Berkshire, bearing in mind Charlie's dictum: "All I want to know is where I'm going to die so I'll never go there." (Inverting really works: Try singing country western songs backwards and you will quickly regain your house, your car and your wife.) If we can't tolerate a possible consequence, remote though it may be, we steer clear of planting its seeds. That is why we don't borrow big amounts and why we make sure that our super-cat business losses, large though the maximums may sound, will not put a major dent in Berkshire's intrinsic value.

在这方面，就像在其他方面一样，我们试图对伯克希尔的未来进行“逆向工程”，牢记查理的格言：“我只想知道我将会死在哪里，这样我就永远不会去那里。”（逆向思维确实有效：试着反向演唱乡村西部歌曲，你会很快重新拥有你的房子、你的车和你的妻子。）如果我们不能容忍可能的后果，无论多么遥远，我们都会避开播下其种子。这就是为什么我们不大量借贷，为什么我们确保，无论最大损失听起来有多大，我们的超级巨灾业务损失都不会对伯克希尔的内在价值造成 major dent。

When we moved to total ownership of GEICO early last year, our expectations were high - and they are all being exceeded. That is true from both a business and personal perspective: GEICO's operating chief, Tony Nicely, is a superb business manager and a delight to work with. Under almost any conditions, GEICO would be an exceptionally valuable asset. With Tony at the helm, it is reaching levels of performance that the organization would only a few years ago have thought impossible.

去年年初我们实现 GEICO 全资控股时，我们的期望很高——而这些期望全部被超越了。从商业和个人角度来看都是如此：GEICO 的运营主管 Tony Nicely 是一位 superb 的企业管理者，与他共事令人愉快。在几乎任何条件下，GEICO 都将是一笔异常有价值的资产。在 Tony 的掌舵下，它正达到组织几年前认为不可能达到的业绩水平。

There's nothing esoteric about GEICO's success: The company's competitive strength flows directly from its position as a low-cost operator. Low costs permit low prices, and low prices attract and retain good policyholders. The final segment of a virtuous circle is drawn when policyholders recommend us to their friends. GEICO gets more than one million referrals annually and these produce more than half of our new business, an advantage that gives us enormous savings in acquisition expenses - and that makes our costs still lower.

GEICO 的成功没有什么深奥之处：公司的竞争优势直接源于其低成本运营商的地位。低成本带来低价格，低价格吸引并留住优质保单持有人。当保单持有人向朋友推荐我们时，就画出了良性循环的最后一段。GEICO 每年获得超过 100 万个推荐，这些推荐产生了我们一半以上的新业务，这一优势为我们在 acquisition 费用上节省了 enormous 资金——并使我们的成本更低。

This formula worked in spades for GEICO in 1996: Its voluntary auto policy count grew 10%. During the previous 20 years, the company's best-ever growth for a year had been 8%, a rate achieved only once. Better yet, the growth in voluntary policies accelerated during the year, led by major gains in the nonstandard market, which has been an underdeveloped area at GEICO. I focus here on voluntary policies because the involuntary business we get from assigned risk pools and the like is unprofitable. Growth in that sector is most unwelcome.

1996 年这一公式对 GEICO 来说非常有效：其自愿汽车保单数量增长了 10%。在过去 20 年中，公司年度最佳增长率是 8%，而且这个 rate 只达到过一次。更好的是，自愿保单的增长在年内加速，主要得益于非标准市场的重大增长，而这一直是 GEICO 一个未开发的领域。我在这里关注自愿保单，是因为我们从指定风险池等获得的非自愿业务是无利可图的。该领域的增长是最不受欢迎的。

GEICO's growth would mean nothing if it did not produce reasonable underwriting profits. Here, too, the news is good: Last year we hit our underwriting targets and then some. Our goal, however, is not to widen our profit margin but rather to enlarge the price advantage we offer customers. Given that strategy, we believe that 1997's growth will easily top that of last year.

如果 GEICO 的增长没有带来合理的承保利润，那将毫无意义。在这方面，消息也是好的：去年我们达到了承保目标，甚至超过了目标。然而，我们的目标不是扩大利润率，而是扩大我们提供给客户的价格优势。基于这一战略，我们相信 1997 年的增长将轻松超过去年。

We expect new competitors to enter the direct-response market, and some of our existing competitors are likely to expand geographically. Nonetheless, the economies of scale we enjoy should allow us to maintain or even widen the protective moat surrounding our economic castle. We do best on costs in geographical areas in which we enjoy high market penetration. As our policy count grows, concurrently delivering gains in penetration, we expect to drive costs materially lower. GEICO's sustainable cost advantage is what attracted me to the company way back in 1951, when the entire business was valued at $7 million. It is also why I felt Berkshire should pay $2.3 billion last year for the 49% of the company that we didn't then own.

我们预计新的竞争对手将进入直销市场，我们一些现有的竞争对手可能会在地理上扩张。尽管如此，我们所享有的规模经济应该能让我们维持甚至扩大围绕我们经济城堡的 protective moat。在我们拥有高市场渗透率的地理区域，我们在成本方面做得最好。随着我们的保单数量增长，同时带来渗透率的提高，我们预计将大幅降低成本。GEICO 可持续的成本优势是早在 1951 年就吸引我关注这家公司的原因，当时整个企业的估值仅为 700 万美元。这也是为什么我觉得伯克希尔去年应该为当时我们尚未拥有的那 49% 的公司支付 23 亿美元。

Maximizing the results of a wonderful business requires management and focus. Lucky for us, we have in Tony a superb manager whose business focus never wavers. Wanting also to get the entire GEICO organization concentrating as he does, we needed a compensation plan that was itself sharply focused - and immediately after our purchase, we put one in.

最大化一家优秀企业的成果需要管理和专注。幸运的是，我们拥有 Tony 这位杰出的管理者，他的业务专注从未动摇。为了让整个 GEICO 组织也能像他一样集中精力，我们需要一个本身具有高度聚焦的薪酬计划——收购后我们立即实施了一个。

Today, the bonuses received by dozens of top executives, starting with Tony, are based upon only two key variables: (1) growth in voluntary auto policies and (2) underwriting profitability on "seasoned" auto business (meaning policies that have been on the books for more than one year). In addition, we use the same yardsticks to calculate the annual contribution to the company's profit-sharing plan. Everyone at GEICO knows what counts.

如今，从 Tony 开始，数十名高管获得的奖金仅基于两个关键变量：(1) 自愿汽车保单的增长，(2)“成熟”汽车业务的承保盈利能力（指记账超过一年的保单）。此外，我们使用相同的标尺来计算公司利润分享计划的年度 contribution。GEICO 的每个人都知道什么是重要的。

The GEICO plan exemplifies Berkshire's incentive compensation principles: Goals should be (1) tailored to the economics of the specific operating business; (2) simple in character so that the degree to which they are being realized can be easily measured; and (3) directly related to the daily activities of plan participants. As a corollary, we shun "lottery ticket" arrangements, such as options on Berkshire shares, whose ultimate value - which could range from zero to huge - is totally out of the control of the person whose behavior we would like to affect. In our view, a system that produces quixotic payoffs will not only be wasteful for owners but may actually discourage the focused behavior we value in managers.

GEICO 的计划体现了伯克希尔的激励薪酬原则：目标应 (1) 针对具体运营业务的经济特性定制；(2) 性质简单，以便易于衡量实现程度；(3) 与计划参与者的日常活动直接相关。作为推论，我们回避“彩票”安排，例如伯克希尔股票期权，其最终价值——可能从零到巨大——完全超脱我们希望影响其行为的人的控制。在我们看来，产生异想天开回报的制度不仅对所有者是浪费，而且实际上可能会抑制我们所珍视的管理者身上的专注行为。

Every quarter, all 9,000 GEICO associates can see the results that determine our profit-sharing plan contribution. In 1996, they enjoyed the experience because the plan literally went off the chart that had been constructed at the start of the year. Even I knew the answer to that problem: Enlarge the chart. Ultimately, the results called for a record contribution of 16.9% ($40 million), compared to a five-year average of less than 10% for the comparable plans previously in effect. Furthermore, at Berkshire, we never greet good work by raising the bar. If GEICO's performance continues to improve, we will happily keep on making larger charts.

每个季度，所有 9,000 名 GEICO 同事都可以看到决定我们利润分享计划 contribution 的结果。1996 年，他们享受了这一过程，因为该计划 literally 超出了年初制定的图表。连我都知道那个问题的答案：扩大图表。最终，结果要求创纪录的 16.9%（4000 万美元）的 contribution，而此前生效的同类计划五年平均值不到 10%。此外，在伯克希尔，我们从不以提高标准来回应 good work。如果 GEICO 的业绩持续改善，我们将乐意继续制作更大的图表。

Lou Simpson continues to manage GEICO's money in an outstanding manner: Last year, the equities in his portfolio outdid the S&P 500 by 6.2 percentage points. In Lou's part of GEICO's operation, we again tie compensation to performance - but to investment performance over a four-year period, not to underwriting results nor to the performance of GEICO as a whole. We think it foolish for an insurance company to pay bonuses that are tied to overall corporate results when great work on one side of the business - underwriting or investment - could conceivably be completely neutralized by bad work on the other. If you bat .350 at Berkshire, you can be sure you will get paid commensurately even if the rest of the team bats .200. In Lou and Tony, however, we are lucky to have Hall-of-Famers in both key positions.

Lou Simpson 继续以 outstanding 的方式管理 GEICO 的资金：去年，他投资组合中的股票跑赢了标准普尔 500 指数 6.2 个百分点。在 Lou 负责的 GEICO 业务部分，我们再次将薪酬与业绩挂钩——但与四年期的投资业绩挂钩，而不是与承保业绩或 GEICO 整体业绩挂钩。我们认为，一家保险公司在业务的某一方面（承保或投资）的杰出工作可能会被另一方面的糟糕工作完全抵消的情况下，将奖金与公司整体业绩挂钩是愚蠢的。如果你在伯克希尔的 batting 0.350，你可以确定你会得到相应的报酬，即使团队其他成员只 batting 0.200。然而，在 Lou 和 Tony 身上，我们幸运地在两个关键职位上都拥有名人堂成员。

Though they are, of course, smaller than GEICO, our other primary insurance operations turned in equally stunning results last year. National Indemnity's traditional business had a combined ratio of 74.2 and, as usual, developed a large amount of float compared to premium volume. Over the last three years, this segment of our business, run by Don Wurster, has had an average combined ratio of 83.0. Our homestate operation, managed by Rod Eldred, recorded a combined ratio of 87.1 even though it absorbed the expenses of expanding to new states. Rod's three-year combined ratio is an amazing 83.2. Berkshire's workers' compensation business, run out of California by Brad Kinstler, has now moved into six other states and, despite the costs of that expansion, again achieved an excellent underwriting profit. Finally, John Kizer, at Central States Indemnity, set new records for premium volume while generating good earnings from underwriting. In aggregate, our smaller insurance operations (now including Kansas Bankers Surety) have an underwriting record virtually unmatched in the industry. Don, Rod, Brad and John have all created significant value for Berkshire, and we believe there is more to come.

尽管它们当然比 GEICO 小，但我们其他的 primary 保险业务去年也同样取得了 stunning 的业绩。National Indemnity 的传统业务综合成本率为 74.2，并且像往常一样，相对于保费规模产生了大量浮存金。在过去三年中，由 Don Wurster 运营的这部分业务平均综合成本率为 83.0。我们的 homestate 业务由 Rod Eldred 管理，尽管吸收了扩展到新州的费用，录得的综合成本率为 87.1。Rod 的三年综合成本率为 amazing 的 83.2。伯克希尔的工人 compensation 业务由 Brad Kinstler 在加州运营，现已扩展到其他六个州，尽管有扩张成本，仍再次实现了优秀的承保利润。最后，Central States Indemnity 的 John Kizer 在实现良好承保收益的同时，创下了保费量的新纪录。总体而言，我们较小的保险业务（现在包括 Kansas Bankers Surety）拥有 virtually 行业内无与伦比的承保记录。Don、Rod、Brad 和 John 都为伯克希尔创造了 significant 价值，我们相信未来还有更多。

In 1961, President Kennedy said that we should ask not what our country can do for us, but rather ask what we can do for our country. Last year we decided to give his suggestion a try - and who says it never hurts to ask? We were told to mail $860 million in income taxes to the U.S. Treasury.

1961 年，肯尼迪总统说过，我们不应该问国家能为我们做什么，而应该问我们能为国家做什么。去年我们决定尝试他的建议——谁说问一问不会有损失？我们被告知要向美国财政部邮寄 8.6 亿美元的所得税。

Here's a little perspective on that figure: If an equal amount had been paid by only 2,000 other taxpayers, the government would have had a balanced budget in 1996 without needing a dime of taxes - income or Social Security or what have you - from any other American. Berkshire shareholders can truly say, "I gave at the office."

对这个数字做一个简单的 perspective：如果只有另外 2,000 名纳税人支付了同等金额，那么政府在 1996 年就能实现预算平衡，而无需来自任何其他美国人的一分钱税收——无论是所得税还是社会保障税等等。伯克希尔的股东可以真正地说：“我在办公室给过了。”

Charlie and I believe that large tax payments by Berkshire are entirely fitting. The contribution we thus make to society's well-being is at most only proportional to its contribution to ours. Berkshire prospers in America as it would nowhere else.

查理和我认为，伯克希尔缴纳大额税款是完全合适的。我们因此对社会福祉所做的贡献，最多只与社会对我们的贡献成比例。伯克希尔在美国繁荣发展，这是在其它任何地方都无法实现的。

The table that follows shows the main sources of Berkshire's reported earnings. In this presentation, purchase-accounting adjustments are not assigned to the specific businesses to which they apply, but are instead aggregated and shown separately. This procedure lets you view the earnings of our businesses as they would have been reported had we not purchased them. For the reasons discussed on pages 65 and 66, this form of presentation seems to us to be more useful to investors and managers than one utilizing generally-accepted accounting principles (GAAP), which require purchase-premiums to be charged off business-by-business. The total earnings we show in the table are, of course, identical to the GAAP total in our audited financial statements.

下表显示了伯克希尔报告收益的主要来源。在此列报中，购买会计调整不分配到具体适用的业务，而是汇总并单独列示。此程序让你能够看到我们业务的收益，就像我们没有购买它们时报告的那样。基于第 65-66 页讨论的原因，我们认为这种列报方式比使用公认会计原则（GAAP）——要求购买溢价按业务逐个冲销——对投资者和经理人更有用。当然，我们在表中显示的总收益与我们经审计财务报表中的 GAAP 总额相同。

(in millions)

（单位：百万美元）

(1) Before the GEICO-related restatement.
(2) Includes R.C. Willey from June 29, 1995.
(3) Includes Helzberg's from April 30, 1995.
(4) Excludes interest expense of Finance Businesses.

(1) 在与 GEICO 相关的重述之前。
(2) 包括自 1995 年 6 月 29 日起的 R.C. Willey。
(3) 包括自 1995 年 4 月 30 日起的 Helzberg's。
(4) 不包括金融业务的利息费用。

In this section last year, I discussed three businesses that reported a decline in earnings - Buffalo News, Shoe Group and World Book. All, I'm happy to say, recorded gains in 1996.

在去年的这一节中，我讨论了三家报告收益下降的企业——布法罗新闻报、鞋业集团和 World Book。我很高兴地说，这三家都在 1996 年录得了增长。

World Book, however, did not find it easy: Despite the operation's new status as the only direct-seller of encyclopedias in the country (Encyclopedia Britannica exited the field last year), its unit volume fell. Additionally, World Book spent heavily on a new CD-ROM product that began to take in revenues only in early 1997, when it was launched in association with IBM. In the face of these factors, earnings would have evaporated had World Book not revamped distribution methods and cut overhead at headquarters, thereby dramatically reducing its fixed costs. Overall, the company has gone a long way toward assuring its long-term viability in both the print and electronic marketplaces.

然而，World Book 并不轻松：尽管该业务已成为国内唯一的百科全书直销商（大英百科全书去年退出了该领域），但其销量还是下降了。此外，World Book 在新 CD-ROM 产品上投入巨资，该产品直到 1997 年初与 IBM 联合推出时才开始产生收入。面对这些因素，如果 World Book 不改进分销方式和削减总部管理费用，从而大幅降低固定成本，收益就会蒸发。总体而言，该公司在确保其在印刷和电子市场的长期 viability 方面已取得长足进步。

Our only disappointment last year was in jewelry: Borsheim's did fine, but Helzberg's suffered a material decline in earnings. Its expense levels had been geared to a sizable increase in same-store sales, consistent with the gains achieved in recent years. When sales were instead flat, profit margins fell. Jeff Comment, CEO of Helzberg's, is addressing the expense problem in a decisive manner, and the company's earnings should improve in 1997.

去年我们唯一的失望是在珠宝领域：Borsheim's 表现不错，但 Helzberg's 的收益出现了 material 下降。其费用水平一直与同店销售额的大幅增长相匹配，这与近年来的增长一致。但当销售持平而不是增长时，利润率下降了。Helzberg's 的 CEO Jeff Comment 正以 decisive 的方式解决费用问题，公司的收益应在 1997 年有所改善。

Overall, our operating businesses continue to perform exceptionally, far outdoing their industry norms. For this, Charlie and I thank our managers. If you should see any of them at the Annual Meeting, add your thanks as well.

总体而言，我们的运营业务继续表现 exceptional，远超出其行业 norm。为此，查理和我感谢我们的经理人。如果你在年会上见到他们中的任何一位，也请一并表达你的感谢。

More information about our various businesses is given on pages 36-46, where you will also find our segment earnings reported on a GAAP basis. In addition, on pages 51-57, we have rearranged Berkshire's financial data into four segments on a non-GAAP basis, a presentation that corresponds to the way Charlie and I think about the company. Our intent is to supply you with the financial information that we would wish you to give us if our positions were reversed.

关于我们各项业务的更多信息在第 36-46 页，你也可以在那里找到我们按 GAAP 报告的部门收益。此外，在第 51-57 页，我们将伯克希尔的财务数据按照非 GAAP 基础重新排列为四个部门，这种列报方式与查理和我思考公司的方式一致。我们的意图是为你提供财务信息，如果我们的位置互换，我们会希望你提供给我们的信息。

Reported earnings are a poor measure of economic progress at Berkshire, in part because the numbers shown in the table presented earlier include only the dividends we receive from investees - though these dividends typically represent only a small fraction of the earnings attributable to our ownership. Not that we mind this division of money, since on balance we regard the undistributed earnings of investees as more valuable to us than the portion paid out. The reason is simple: Our investees often have the opportunity to reinvest earnings at high rates of return. So why should we want them paid out?

报告的收益是衡量伯克希尔经济进展的一个 poor measure，部分原因前面表格中显示的数字只包括我们从被投资公司收到的股息——尽管这些股息通常仅代表归因于我们所有权的收益的一小部分。并不是我们介意这种 money 的划分，因为总的来说，我们认为被投资公司的未分配收益对我们来说比支付的部分更有价值。原因很简单：我们的被投资公司 often 有机会以高回报率 reinvest 收益。那么我们为什么希望它们被支付出来呢？

To depict something closer to economic reality at Berkshire than reported earnings, though, we employ the concept of "look-through" earnings. As we calculate these, they consist of: (1) the operating earnings reported in the previous section, plus; (2) our share of the retained operating earnings of major investees that, under GAAP accounting, are not reflected in our profits, less; (3) an allowance for the tax that would be paid by Berkshire if these retained earnings of investees had instead been distributed to us. When tabulating "operating earnings" here, we exclude purchase-accounting adjustments as well as capital gains and other major non-recurring items.

然而，为了描绘比报告收益更接近伯克希尔经济 reality 的情况，我们采用了“透视”收益的概念。我们计算这些收益时，它们包括：(1) 上一节报告的营业收益，加上；(2) 我们在主要被投资公司保留的营业收益中所占的份额（根据 GAAP 会计，这些未反映在我们的利润中），减去；(3) 如果这些被投资公司的保留收益分配给我们，伯克希尔将支付的税款 allowance。在 tabulating“营业收益”时，我们排除了购买会计调整以及资本利得和其他主要的非经常性项目。

The following table sets forth our 1996 look-through earnings, though I warn you that the figures can be no more than approximate, since they are based on a number of judgment calls. (The dividends paid to us by these investees have been included in the operating earnings itemized on page 12, mostly under "Insurance Group: Net Investment Income.")

下表列出了我们 1996 年的透视收益，不过我警告你，这些数字只能是近似值，因为它们基于许多判断。（这些被投资公司支付给我们的股息已包含在第 12 页分项列出的营业收益中，主要是在“保险集团：净投资收益”项下。）

(1) Does not include shares allocable to minority interests
(2) Calculated on average ownership for the year
(3) The tax rate used is 14%, which is the rate Berkshire pays on the dividends it receives

(1) 不包括可分配给少数股东权益的股份
(2) 按年度平均持股计算
(3) 使用的税率为 14%，即伯克希尔对收到的股息支付的税率

Below we present our common stock investments. Those with a market value of more than $500 million are itemized.

以下是我们普通股投资的情况。市值超过 5 亿美元的投资会逐项列出。

12/31/96

1996 年 12 月 31 日

• Represents tax-basis cost which, in aggregate, is $1.2 billion less than GAAP cost.

• 代表税基成本，总计比 GAAP 成本低 12 亿美元。

Our portfolio shows little change: We continue to make more money when snoring than when active.

我们的投资组合变化不大：我们继续在打盹时比在活动时赚更多的钱。

Inactivity strikes us as intelligent behavior. Neither we nor most business managers would dream of feverishly trading highly-profitable subsidiaries because a small move in the Federal Reserve's discount rate was predicted or because some Wall Street pundit had reversed his views on the market. Why, then, should we behave differently with our minority positions in wonderful businesses?

不活动在我们看来是明智的行为。无论是我们还是大多数企业管理者，都不会因为预测美联储贴现率的小幅变动，或者因为某个华尔街权威人士改变了对市场的看法，就梦想要狂热地买卖高利润的子公司。那么，为什么我们在优秀企业的少数股权上就应该有不同的行为呢？

Warren E. Buffett
February 28, 1997
Chairman of the Board

沃伦·E·巴菲特
1997 年 2 月 28 日
董事会主席